Project Summary Autism Spectrum Disorders (ASD) comprise a range of heterogeneous neurodevelopmental disorders that manifest in early childhood with abnormal motor and social behaviors. Autism has long been synonymous with social reward processing deficits and repetitive behaviors, but has recently been associated with deficits in goal-directed decision making and the processing of non-social rewards as well. Both the complex genetic etiology and variable phenotypic presentation of ASD have made an understanding of the specific higher-order circuit dysfunctions mediating this complex behavioral picture difficult to ascertain. With this proposal, we seek to explore how perturbations to one high-risk ASD gene, Neurexin1?, contribute to specific circuit disruptions that mediate the processing of rewards. Preliminary data from goal-directed tasks developed in our lab suggest abnormalities in non-social reward processing in Neurexin1? mutant mice. Mice exhibit a deficit in flexible response behavior in new reward contingencies that is more pronounced in reward environments with larger disparities in net benefit. While recent studies have begun to explore both the role of the dorsomedial striatum in flexible goal-directed behaviors, and the role of projection inputs onto striatal compartments in regulating their activity, few have mapped an association between genetic insults to these striatal circuits and behavioral pathology in neuropsychiatric disease. I hypothesize that Neurexin1? loss-of-function disturbs particular afferent projection inputs innervating the dorsomedial striatum, and that these circuit abnormalities mediate specific features of the goal-directed phenotypes we observe in Neurexin1? mutant mice. In Aim 1, I will use modified versions of our behavioral task to further refine our understanding of how Neurexin1? mutants weigh benefits and costs to execute value-based control of motor output. In Aim 2, I will perform circuit-specific knockouts of Neurexin1? using a conditional knockout line and utilize our behavioral paradigms to assess the relative contribution of these targeted molecular disruptions to the observed goal-directed phenotypes. This approach will serve as the foundation for a circuit-based analysis of altered neuronal coding that occurs in Neurexin1? mutants during value-based choice. In sum, I believe these aims should provide an excellent opportunity for training in cutting-edge behavioral and molecular neuroscience approaches, and have the potential to greatly inform our understanding of the cellular basis of reward processing dysfunction in neuropsychiatric disease. 1